Walid, I just wanted to mention genetic algorithms; control engineers use them e.g. in chemical plants to dynamically optimize variable inputs (pressure in a pipe, temperature, ingredient mixes...) for measurable outputs (valuable gunk at the end), the point being that the inter-relations among all the inputs, at the level of precision possible to the machinery, is too esoteric for an exact qualitative model.
I think that managing diverse simultaneous applications (that each flourish in their own mix of bandwidth, cache, numbers of threads on cores &c) where a mix of hardware is available in a heterogeneous cluster, might be a similar challenge. So if someone were interested in trying a GA for dynamic performance optimization of long-running applications with mixed available hardware I'd be glad to hear of it and I'd want to help. Peter On Sun, Jan 6, 2013 at 8:38 AM, Walid <[email protected]> wrote: > Dear All, > > At work we are starting to evaluate Configuration management to be used to > manage several diverse hpc clusters, and their diverse node types. I wanted > to see what are other admins, and HPC users experience like, the ones that > we will start evaluating are CFEngine3, Puppet, Chef, Saltstack, ansible, > and blueprint. there might other products that we need to evluate in > partnership such as Foreman, spacewalk, ..etc. > > I would like to hear from you if you did evaluate such tools, or using > one, or have a different strategy in keeping and maintaining > configurations. > > Thank you, > > Walid > > _______________________________________________ > Beowulf mailing list, [email protected] sponsored by Penguin Computing > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > >
_______________________________________________ Beowulf mailing list, [email protected] sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
